Segmentation Methodologies of Diabetic Foot Ulcer Images

Categories: Diabetes

C# and Umadevi. P #Professor, Centre for Research and Development, Department of Electronics andCommunication Engineering#PG scholar, Department of Electronics and Communication Engineering, KPR Institute of Engineering and Technology,Coimbatore " 641 407, Tamil Nadu, INDIA,ABSTRACT:Diabetic foot ulcer (DFU) is the complication of Diabetic mellitus (DM) where 15% develop foot ulcer and 1/4 will leads to amputation. Breaking of neuropathic (nerves) and weakening of vascular (blood vessels) are the two ways of diabetics. Extraordinary increase of glucose in the blood results in nerve injury, termed as diabetic neuropathy.

Blood sugar control, latest dressing methodologies, wound debridement and offloading modalities are some of the factors of DFU management to heal the wounded area at the starting stage where the patient with more affected area are been healed by wound segmentation process. Segmentation is one of the major steps in medical image processing that divides a given image into multiple regions to analyze and distinguish into different objects in the region of interest. There is different image segmentation techniques have been developed in order to make the images appropriate to evaluate.

Get quality help now
WriterBelle
WriterBelle
checked Verified writer

Proficient in: Diabetes

star star star star 4.7 (657)

“ Really polite, and a great writer! Task done as described and better, responded to all my questions promptly too! ”

avatar avatar avatar
+84 relevant experts are online
Hire writer

This paper presents a survey on schematic approach for processing and segmentation methodologies that can be applied to diabetic foot ulcer images for better clinical diagnosis.

Keywords: Diabetic Foot Ulcer, DFU image preprocessing, image segmentation, clinical diagnosis.INTRODUCTION:Some pre-processing techniques where used in image processing such as image enhancement used to remove noise or correct the contrast in the image, the thresholding to remove thebackground containing any scenes, watermark and noise. Some non-uniform illumination are been removed by morphological operation [1].

Get to Know The Price Estimate For Your Paper
Topic
Number of pages
Email Invalid email

By clicking “Check Writers’ Offers”, you agree to our terms of service and privacy policy. We’ll occasionally send you promo and account related email

"You must agree to out terms of services and privacy policy"
Write my paper

You won’t be charged yet!

Some image database is created with a live scan, of two techniques which are combined together and structured pre-processing steps are proposed which encompasses

  • the removal of non-illumination
  • contrast enhancement
  •  binarisation by thresholding,
  •  median filtering and
  •  histogram equalized FFT enhancement and weighting vector is carried[2][3].

Finally segmentation techniques of image are used to separate graphics. Segmentation which are still being used for research purpose are Modified Chain vase algorithm, Accelerated mean shift algorithm, Two-tier transfer learning, Thresholding technique, K-means algorithm. Images are divided into two types on the basis of color, as gray scale and color images. Therefore image segmentation for color images is totally different from gray scale images, for content based image retrieval [4]. The property of a pixel in an image and information of pixels near to that pixel are two basic parameters for any image segmentation algorithm. It can also be representing as similarity of pixels in any region and discontinuity of edges in image. Modified Chain Vase algorithm is been formulated by the body temperature which should not exceed 2.2celsis, analyzed by IR thermal image processing. Accelerated mean shift is used to segment the wound area by color coding the skin feature.

Two- tier transfer learning is used to automatically segment the area [5]. The result taken from image segmentation process; is used to determine the quality of further process. Image segmentation algorithms play an important role in medical applications, i.e., diagnosis of diseases related to brain [3] heart, knee, spine, pelvis, prostate and blood vessel, and pathology localization. Therefore, Image segmentation is still a very hot area for image processing field. Image segmentation is also used to differentiate different objects in the image, and divided into foreground and background, where foreground of image is related to the region of interest, and background is the rest of the image [6]. Hence, image segmentation will separate these two parts from one another.

MANAGEMENT ON DIABETIC FOOT ULCER

However, numerous studies have shown that proper management of DFU can greatly reduce the complication and the severity of the wound. Segmentation plays an important role in differentiating the wound area in different objects to examine clearly.DFU can be identified and segmented by color feature analysis caused by low nutrients and pressure on the foot.

LITERATURE REVIEW OF IMAGE SEGMENTATION TECHNIQUES

Diabetic foot ulcer on a patient is been identified by several process of implementation. Before segmentation identification is carried over for identifying the damaged tissue and the location of the nerve damage by SVM based identification and foot anthropometry by using hyper spectral imaging process [7][8]. Segmentation on foot ulcer is been carried over by several process which is used to determine the area of the wound by color feature analysis which is been caused by low nutrients, pressure on the foot and low nutrients on the affected tissue.a) Modified Chain Vase Algorithm:On 2017, December Modified Chan vase algorithm [9] method named global Region Based chain Vase Algorithm is been implemented to detect foot ulcer by body temperature and analyzed by IR thermal image processing and several iteration process is carried out and ROI is measured. The edges of foot images is taken by the infrared camera where the segmentation boundary implicit with the level set function so the initial curve is placed on the image.

The modified chain vase is based on morphological processing where the temperature is measured on the risky area of the foot. In this process the input image is been processed by different types of masks and initiate Chain-Vase vector to start the iteration. The iteration is carried over until the desired results are obtained. After the iteration process in the global region based segmentation the pale pale is left in the foot, where remaining foot may be damaged that part is eliminated to save the entire foot.

Accelerated Mean Shift Algorithm:On 2016Accelerated mean shift algorithm is used to accelerate the wound area and by using color coding feature the boundary area is detected, further MATLAB application is used to process the captured image where color based segmentation is used to detect skin color, wound boundary [10]. The mean-shift filtering algorithm is been suitable for parallel implementation, where it belongs to the density estimation process based non parametric clustering methods, in which the feature space can be considered as the empirical probability density function of the parameter. In general, the mean-shift algorithm models the feature vectors associated with each pixel (e.g., color and position in the image grid) as samples from an unknown probability density function f(x) and then finds clusters in n number of distribution [11].

Another method is based on automatic assessment of diabetic foot ulcer based on wound area determination, colorsegmentation is processed by wound image assessment algorithm where image capture box is used to capture the image by warm light LED light which is compact, inexpensive or by using MATLAB application and the color segmentation is carried by accelerated mean shift algorithmprocess [12] [13]Image SegmentationTechniquesModified ChainVase AlgorithmAccelerated MeanShift AlgorithmTwo-Tier transfer learninghresholdingTechniqueK-meansAlgorithmFig:1

IMAGE SEGMENTATION TECHNIQUES

Two-Tier Transfer Learning:Segmentation on diabetic foot ulcer is carried by a two-tier transfer learning to train the fully Convolutional networks (FCNs) to automatically segment the ulcer and surrounding skin[14], the dataset is been annotated by growth truth model which is captured by Nikon D3300 on close range focus to avoid the blurriness. The proposed two-tier transfer learning FCN models achieve a dice similarity coefficient of 0.794 for ulcer region, 0.851 for surrounding skin region and 0.899 for the combination of both regions as such FCN Alex net, FCN 32 (32*32),FCN 16 (16*16),FCN 8 (8*8) pixel formatted. FCN Alex net classifies different objects of classes on the image net datasets, where another three combination are customized by up sampling process. The comparison is made where FCN-16 and FCN-8 have low level features which produce more accurate segmentation.

Thresholding Technique:The foot ulcer detection using image processing technique is used to determine the area and perimeter of the damaged wound by thresholding technique and edge detection. Various preprocessing technique is carried which convert into gray scale image. To enhance the contrast the gray scale filtering and image enhancement is applied. In image segmentation the preprocessed image is taken, and the ulcer in the image is segmented separately by thresholding technique and edge detection technique. The foot ulcer is segmented and morphology of image is restored then the area and perimeter of the image is calculated. This technique is applied two times for a one patient to compare the result of privies visit and the next visit.

Based on this, the area of ulcer is healing or not healing can be determined. Thus thresholding technique improves the performance of determining the area and the perimeter of the wounded area by area function and the perimeter function to detect and segment the record [15]. Various function of morphological operation is used for accurate result, finally the output of closed operation is have some holes in the image so by using hole filling technique the image will become so clear. Then by using remove operation we can remove the other parts of unwanted information present image. K-Means Algorithm:The design of Smartphone-Based Wound Assessment System for foot ulcer [16] is designed by K-means algorithm which is more accurate for clustering where the input image is captured by smart phone a lab setup is made for evaluation. The captured image is transformed by mobile application to the server and to the analyzer is done over by using smart phone application where the captured image is transformed by mobile application to the server and to the analyzer which is done by K-means algorithm [17] the image is clustered and the output is more accurate when compare to those algorithms.

The K-means algorithm is most well-known in the partitioned clustering. K-means algorithm decide the numbers of the cluster and then choose randomly data points (pixels or image) in the whole image as the centroids in clusters, then find out nearest centroid of every single data point (pixel or image) and classify the data point into that cluster the centroid located and all data points are classified in some cluster. Calculate the centroid of every cluster. Using the K-means algorithm, it has an advantage of less computing time. In other words, the partitional clustering is faster than the hierarchical clustering [18].Segmentation process on color analysis is carried by Red-Yellow-Black- White (RYKW) probability map where yellow and black probability maps are designed to handle the granulation, slough and eschar tissues, and the white probability map is to detect the white label card for measuring the area affected[19].

CONCLUSION

The overall existing determines the area, size, texture, detection, color coding and segmentation is processed for different algorithm. Various analyses are carried over such as cell separation and quantification, bone separation, trabecular bone analysis, tissue texture separation, segmentation through classification. Through the study of existing process on diabetic foot ulcer segmentation on image is most widely used where different algorithms are examined and verified with large datasets where K-Means produce an accurate result on segmentation when compared to the other algorithm. In this evaluation of main image segmentation techniques are used for the purpose of image analysis. Hence no perfect method for image segmentation is used because the result of image segmentation is depends on many factors, i.e., pixel color, texture, intensity, similarity of images, image content, and problem domain. Therefore, it is not possible to consider a single method for all type of images nor all methods can perform well for a particular type of image. Hence, it is good to use hybrid solution consists of multiple methods for image segmentation problem.

REFERENCE:

  1. C. Karthikeyini Chockaian, Rajamani Vayanaperumal, Bommanna Raja Kanagaraj Newapproach for identifying hereditary relation using primary fingerprint patterns. Vol 7, Issue 5July 2013, DOI: 10.1049/iet-ipr.2012.0399.
  2.  C. Karthikeyini, V. Rajamani, K. Bommanna Raja, Exploring fingerprints using composite minutiae descriptors to determine hereditary relation across multiple generations, International Journal On Biomedical Engineering and Technology, Vol. 15, No. 3, 2014.
  3.  C. Karthikeyini , K. Bommanna Raja , Dr. M. Madheshwaran Study on UltrasoundKidney Images Using Principal Component Analysis: A Preliminary Result.[4] M. Yasmin, S. Mohsin, I. Irum, and M. Sharif, "Content based image retrieval by shape, color and relevance feedback," Life Science Journal, vol. 10, 2013.
  4.  W. X. Kang, Q. Q. Yang, and R. P. Liang, "The comparative research on image segmentation algorithms," in Proc. First International Workshop on Education Technology and Computer Science, 2009. ETCS'09. pp. 703-707, 2009.
  5. [6] M. Yasmin, S. Mohsin, M. Sharif, M. Raza, and S. Masood, "Brain Image Analysis: A Survey," World Applied Sciences Journal, vol. 19, pp. 1484-1494, 2012.
  6. H. Hedberg, "A survey of various image segmentation techniques," Dept. of Electroscience, Box, vol. 118, 2010
  7. Sorna Percy.G, Anumuga Maria Devi An efficiently identify the diabetic foot ulcer based on foot anthropometry using hyper spectral imaging2016-International Journal Of Information Technology And Management Information System. Volume 7, Issue 2, May-August-2016.
  8.  Lei.Wang,Peder, Emmanuel Agu, Diane Strong Area determination of diabetic foot ulcer image using a cascaded two stage SVM based classification 2016-IEEE Transaction in biomedical engineering. Volume 64, Issue 9(23rd Nov)DOI:10.1109/TBME.2016.2632522.
  9.  Shaik.Bajid.vali, Dr.Anil Kumar Sharma, Dr.Syed Musthak Ahmed Implementation of modified chan vase algorithm to detect and analyze diabetic foot ulcer 2017-International Journal on Recent Trends In Electrical Electronics And Computing Technology. Dec-11,DOI:201710.1109/icrteect2017.25.
  10.  M.Saratha, V.Mohana Priya Detection of diabetic wounds based on segmentation using accelerated mean shift algorithm 2016-International Journal Of Advanced Research In Computer Science And Software Engineering,Volume 6, Issue 2, February 2016.
  11.  Lei.Wang Peder, C.Pedersen, Bengisu Tulu, Diane.M.Strong Automatic assessment system of diabetic foot ulcer based on wound area determination, color segmentation and healing score evaluation SAGE 2016-International Journal Of Diabetics science And Technology, Volume 10, DOI:10.1177/1932296815599004.
  12. C.Geetha, Sathish Kumar Wound assessment for diabetic patients using MATLAB application in smart phone 2016-International Journal Of Advanced Research Trends In Engineering And Technology,Volume 3, Special Issue 24, April 2016.[14] Sadhana.S.Jadhav, Manisha.H. Naog Hare A survey on wound assessment system patients of foot ulcer diabetic identification based on smart phone 2015-International Journal Of Innovative Research In Computer And Communication Engineering,Volume. 3, Issue 11, November 2015.
  13.  Manu Goyal Neil, D. Reevesy Satyan, Rajbhandariz Jennifer, Spraggx Moi Hoon Yap Fully convolutional networks for diabetic foot ulcer segmentation 2017-IEEE International Conference On Systems, Man, And Cybernetics (SMC) Dec-1,DOI:201710.1109/smc.2017.812265.
  14.  Hari Prasad.R, Sharmila.N Foot ulcer detection using image processing technique 2016- International Journal Of Computing and Technology, Volume 3, Issue 3, March 2016.
  15. Vidyashree Dalya , Dr. D.K. Shedge Design of Smartphone-Based Wound Assessment System 2016 International Conference on Automatic Control and Dynamic Optimization Techniques (ICACDOT) 16,March 2017,DOI:10.1109/ICACDOT.2016.7877679.
  16.  Sadhana.S.Jadhav, Manisha.H. Naog Hare A survey on wound assessment system patients of foot ulcer diabetic identification based on smart phone 2015-International Journal Of Innovative Research In Computer And Communication Engineering,Volume. 3, Issue 11, November 2015.
  17.  C.Geetha, Sathish Kumar Wound assessment for diabetic patients using MATLAB application in smart phone 2016-International Journal Of Advanced Research Trends In Engineering And Technology,Volume 3, Spcial Issue 24, April 2016.
  18. Mohammad FaizalAhmad, FauIbrahim Khansa Karen, Catignani Gayle Gordillo Computerized segmentation and measurement of chronic wound images 2015-ELSEVIER DOI:10.1016/j.compbiomed.2015.02.015 0010-4825.[21] [22]
Updated: Sep 25, 2020
Cite this page

Segmentation Methodologies of Diabetic Foot Ulcer Images. (2019, Aug 20). Retrieved from https://studymoose.com/segmentation-methodologies-of-diabetic-foot-ulcer-images-essay

Segmentation Methodologies of Diabetic Foot Ulcer Images essay
Live chat  with support 24/7

👋 Hi! I’m your smart assistant Amy!

Don’t know where to start? Type your requirements and I’ll connect you to an academic expert within 3 minutes.

get help with your assignment